Phi-Former: A Pairwise Hierarchical Approach for Compound-Protein Interactions Prediction

TL;DR

Phi-former introduces a hierarchical pairwise learning approach that models compound-protein interactions at multiple biological levels, improving prediction accuracy and interpretability for drug discovery.

Contribution

It is the first to incorporate motif-level biological recognition into hierarchical pairwise CPI modeling, enhancing both accuracy and interpretability.

Findings

Achieves superior performance on CPI prediction tasks.

Provides interpretable insights into atom and motif activation.

Supports rational drug design and precision medicine.

Abstract

Drug discovery remains time-consuming, labor-intensive, and expensive, often requiring years and substantial investment per drug candidate. Predicting compound-protein interactions (CPIs) is a critical component in this process, enabling the identification of molecular interactions between drug candidates and target proteins. Recent deep learning methods have successfully modeled CPIs at the atomic level, achieving improved efficiency and accuracy over traditional energy-based approaches. However, these models do not always align with chemical realities, as molecular fragments (motifs or functional groups) typically serve as the primary units of biological recognition and binding. In this paper, we propose Phi-former, a pairwise hierarchical interaction representation learning method that addresses this gap by incorporating the biological role of motifs in CPIs. Phi-former represents compounds and proteins hierarchically and employs a pairwise pre-training framework to model interactions systematically across atom-atom, motif-motif, and atom-motif levels, reflecting how biological systems recognize molecular partners. We design intra-level and inter-level learning pipelines that make different interaction levels mutually beneficial. Experimental results demonstrate that Phi-former achieves superior performance on CPI-related tasks. A case study shows that our method accurately identifies specific atoms or motifs activated in CPIs, providing interpretable model explanations. These insights may guide rational drug design and support precision medicine applications.